1. 23 Jan, 2013 40 commits
    • wangxianzhu@chromium.org's avatar
      Should update compositing state when an out-of-view fixed position element becomes in-view · a973b01a
      wangxianzhu@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107410
      
      Reviewed by Simon Fraser.
      
      Source/WebCore:
      
      When a non-compositing page contains some not-composited fixed position element because of bounds out-of-view, when the element changes position and RenderLayerCompositor::updateCompositingLayers() is called, the function may return early because of "if (!m_reevaluateCompositingAfterLayout && !m_compositing)" without updating the compositing layers.
      
      Set m_reevaluateCompositingAfterLayout when a fixed position element is not composited because of bounds out-of-view.
      
      Test: compositing/layer-creation/fixed-position-change-out-of-view-in-view.html
      
      * rendering/RenderLayerCompositor.cpp:
      (WebCore::RenderLayerCompositor::requiresCompositingForPosition):
      
      LayoutTests:
      
      Tests if compositing status is properly updated when fixed position elements change from out-of-view to in-view and back.
      
      * compositing/layer-creation/fixed-position-change-out-of-view-in-view-expected.txt: Added.
      * compositing/layer-creation/fixed-position-change-out-of-view-in-view.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140593 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a973b01a
    • ap@apple.com's avatar
      Build fix. · ec1703e3
      ap@apple.com authored
              * Platform/CoreIPC/Connection.cpp: (CoreIPC::Connection::sendSyncMessage):
              Copy/paste trouble fix.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140591 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ec1703e3
    • tony@chromium.org's avatar
      Unreviewed, set svn:eol-style to CRLF on Windows .sln files. · 71b603a3
      tony@chromium.org authored
      Source/JavaScriptCore: 
      
      * JavaScriptCore.vcproj/JavaScriptCore.sln: Modified property svn:eol-style.
      * JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: Modified property svn:eol-style.
      
      Source/ThirdParty: 
      
      * gtest/msvc/gtest-md.sln: Added property svn:eol-style.
      * gtest/msvc/gtest.sln: Added property svn:eol-style.
      
      Source/WebCore: 
      
      * WebCore.vcproj/WebCore.sln: Modified property svn:eol-style.
      * WebCore.vcproj/WebCore.submit.sln: Modified property svn:eol-style.
      
      Source/WebKit: 
      
      * WebKit.vcxproj/WebKit.sln: Added property svn:eol-style.
      
      Source/WebKit/win: 
      
      * WebKit.vcproj/WebKit.sln: Modified property svn:eol-style.
      * WebKit.vcproj/WebKit.submit.sln: Modified property svn:eol-style.
      
      Source/WTF: 
      
      * WTF.vcproj/WTF.sln: Added property svn:eol-style.
      
      Tools: 
      
      * CLWrapper/CLWrapper.sln: Modified property svn:eol-style.
      * DumpRenderTree/DumpRenderTree.sln: Modified property svn:eol-style.
      * MIDLWrapper/MIDLWrapper.sln: Modified property svn:eol-style.
      * WebKitTestRunner/WebKitTestRunner.sln: Modified property svn:eol-style.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140590 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      71b603a3
    • abarth@webkit.org's avatar
      BackgroundHTMLParser::sendTokensToMainThread should use bind · 1da05111
      abarth@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107637
      
      Reviewed by Eric Seidel.
      
      Source/WebCore: 
      
      This patch replaces our hand-written implementation of bind for
      didReceiveTokensFromBackgroundParser with bind from Functional.h. To
      use the generic version of bind, we need to switch to using WeakPtr to
      hold a reference to the main thread parser in the BackgroundHTMLParser.
      
      * html/parser/BackgroundHTMLParser.cpp:
      (WebCore::BackgroundHTMLParser::BackgroundHTMLParser):
      (WebCore::BackgroundHTMLParser::sendTokensToMainThread):
      (WebCore::BackgroundHTMLParser::createPartial):
      * html/parser/BackgroundHTMLParser.h:
      (WebCore::BackgroundHTMLParser::create):
      (BackgroundHTMLParser):
      (ParserMap):
      * html/parser/HTMLDocumentParser.cpp:
      (WebCore::HTMLDocumentParser::HTMLDocumentParser):
      (WebCore::HTMLDocumentParser::startBackgroundParser):
      (WebCore::HTMLDocumentParser::stopBackgroundParser):
      * html/parser/HTMLDocumentParser.h:
      (HTMLDocumentParser):
      
      Source/WTF: 
      
      * wtf/Functional.h:
          - I had to re-work the approach to validating WeakPtr |this|
            arguments a bit. Previously you couldn't pass a WeakPtr as a
            non-|this| argument to a function because we would try to unwrap
            it into a raw pointer.
      * wtf/WeakPtr.h:
      (WTF::WeakPtrFactory::revokeAll):
      (WeakPtrFactory):
          - Let clients revoke all outstanding WeakPtrs without needing to
            destroy the WeakPtrFactory.
      * wtf/chromium/MainThreadChromium.cpp:
      (WTF::callFunctionObject):
      (WTF::callOnMainThread):
          - Implement callOnMainThread for Function objects. The non-Chromium
            implementation of MainThread.cpp already implements this feature.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140589 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1da05111
    • roger_fong@apple.com's avatar
      Unreviewed. Cleanup VS2010 WebCore project. · 1bcd5930
      roger_fong@apple.com authored
      Lots of files that no longer exist in solution.
      
      * WebCore.vcxproj/WebCore.vcxproj:
      * WebCore.vcxproj/WebCore.vcxproj.filters:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140588 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1bcd5930
    • pilgrim@chromium.org's avatar
      [Chromium] Remove WebArrayBufferView and WebSerializedScriptValue from chromium/public/platform/ · 8862617d
      pilgrim@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107720
      
      Reviewed by Darin Fisher.
      
      These headers have been moved to chromium/public/ and all
      references downstream were changed in
      https://codereview.chromium.org/11946050/.
      No DEPS roll necessary because it's already above 177695.
      
      * WebKit.gyp:
      * public/platform/WebArrayBufferView.h: Removed.
      * public/platform/WebSerializedScriptValue.h: Removed.
      * src/DeliveredIntentClientImpl.cpp:
      * src/IDBCallbacksProxy.cpp:
      * src/IDBCursorBackendProxy.cpp:
      * src/WebArrayBufferView.cpp:
      * src/WebBindings.cpp:
      * src/WebDOMCustomEvent.cpp:
      * src/WebDOMMessageEvent.cpp:
      * src/WebFrameImpl.cpp:
      * src/WebHistoryItem.cpp:
      * src/WebIDBCallbacksImpl.cpp:
      * src/WebIntent.cpp:
      * src/WebIntentRequest.cpp:
      * src/WebSerializedScriptValue.cpp:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140586 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8862617d
    • arv@chromium.org's avatar
      Layout Test platform/chromium-win/fast/events/panScroll-panIcon.html is failing · ce2adbbf
      arv@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107729
      
      Unreviewed gardening.
      
      * platform/chromium/TestExpectations:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140585 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ce2adbbf
    • oliver@apple.com's avatar
      Replace numerous manual CRASH's in JSC with RELEASE_ASSERT · 51b41da2
      oliver@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=107726
      
      Reviewed by Filip Pizlo.
      
      Fairly manual change from if (foo) CRASH(); to RELEASE_ASSERT(!foo);
      
      * assembler/MacroAssembler.h:
      (JSC::MacroAssembler::branchAdd32):
      (JSC::MacroAssembler::branchMul32):
      * bytecode/CodeBlockHash.cpp:
      (JSC::CodeBlockHash::CodeBlockHash):
      * heap/BlockAllocator.h:
      (JSC::Region::create):
      (JSC::Region::createCustomSize):
      * heap/GCAssertions.h:
      * heap/HandleSet.cpp:
      (JSC::HandleSet::visitStrongHandles):
      (JSC::HandleSet::writeBarrier):
      * heap/HandleSet.h:
      (JSC::HandleSet::allocate):
      * heap/Heap.cpp:
      (JSC::Heap::collect):
      * heap/SlotVisitor.cpp:
      (JSC::SlotVisitor::validate):
      * interpreter/Interpreter.cpp:
      (JSC::Interpreter::execute):
      * jit/ExecutableAllocator.cpp:
      (JSC::DemandExecutableAllocator::allocateNewSpace):
      (JSC::ExecutableAllocator::allocate):
      * jit/ExecutableAllocator.h:
      (JSC::roundUpAllocationSize):
      * jit/ExecutableAllocatorFixedVMPool.cpp:
      (JSC::FixedVMPoolExecutableAllocator::FixedVMPoolExecutableAllocator):
      (JSC::ExecutableAllocator::allocate):
      * runtime/ButterflyInlines.h:
      (JSC::Butterfly::createUninitialized):
      * runtime/Completion.cpp:
      (JSC::evaluate):
      * runtime/JSArray.h:
      (JSC::constructArray):
      * runtime/JSGlobalObject.cpp:
      (JSC::slowValidateCell):
      * runtime/JSObject.cpp:
      (JSC::JSObject::enterDictionaryIndexingModeWhenArrayStorageAlreadyExists):
      (JSC::JSObject::createArrayStorage):
      * tools/TieredMMapArray.h:
      (JSC::TieredMMapArray::append):
      * yarr/YarrInterpreter.cpp:
      (JSC::Yarr::Interpreter::allocDisjunctionContext):
      (JSC::Yarr::Interpreter::allocParenthesesDisjunctionContext):
      (JSC::Yarr::Interpreter::InputStream::readChecked):
      (JSC::Yarr::Interpreter::InputStream::uncheckInput):
      (JSC::Yarr::Interpreter::InputStream::atEnd):
      (JSC::Yarr::Interpreter::interpret):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140584 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      51b41da2
    • jchaffraix@webkit.org's avatar
      [CSS Grid Layout] Add support for max-content · d5bbdab2
      jchaffraix@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107604
      
      Reviewed by Tony Chang.
      
      Source/WebCore:
      
      Tests: fast/css-grid-layout/minmax-max-content-resolution-columns.html
             fast/css-grid-layout/minmax-max-content-resolution-rows.html
      
      This change implements max-content on top of the infrastructure introduced as part
      of implementing min-content (bug 106474). No effort was made to share code, which
      is what was done for min-content. The sharing will occur in follow-up refactoring(s)
      to benefit from the extra testing but also the extra code to make the direction more
      obvious.
      
      * rendering/RenderGrid.cpp:
      (WebCore::RenderGrid::maxContentForChild):
      Added this helper, similar to minContentForChild.
      
      (WebCore::RenderGrid::resolveContentBasedTrackSizingFunctions):
      Patched the function to handle max-content per the specification's algorithm.
      
      * rendering/RenderGrid.h:
      Added maxContentForChild.
      
      LayoutTests:
      
      * fast/css-grid-layout/minmax-max-content-resolution-columns-expected.txt: Added.
      * fast/css-grid-layout/minmax-max-content-resolution-columns.html: Added.
      * fast/css-grid-layout/minmax-max-content-resolution-rows-expected.txt: Added.
      * fast/css-grid-layout/minmax-max-content-resolution-rows.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140583 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      d5bbdab2
    • ap@apple.com's avatar
      <rdar://problem/13064414> REGRESSION(r139514):... · d4a195ac
      ap@apple.com authored
              <rdar://problem/13064414> REGRESSION(r139514): didFailToSendSyncMessage is not called by Connection
              https://bugs.webkit.org/show_bug.cgi?id=107715
      
              Reviewed by Anders Carlsson.
      
              * Platform/CoreIPC/Connection.cpp:
              (CoreIPC::Connection::sendSyncMessage):
              (CoreIPC::Connection::sendSyncMessageFromSecondaryThread):
              Correcting a misapplied edit - didFailToSendSyncMessage should be kept in main
              thread message sending, not in secondary thread.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140582 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      d4a195ac
    • arv@chromium.org's avatar
      [chromium] Layout test rebaselines · a39ff63b
      arv@chromium.org authored
      Unreviewed rebaseline
      
      * platform/chromium-win-xp/media/media-can-play-ogg-expected.txt: Added.
      * platform/chromium-win/media/media-can-play-ogg-expected.txt: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140581 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a39ff63b
    • mark.lam@apple.com's avatar
      Assert that Supplementable objects is only used in their creator thread. · 570be473
      mark.lam@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=107717.
      
      Reviewed by Adam Barth.
      
      No new tests.
      
      * platform/Supplementable.h:
      (WebCore):
      (WebCore::Supplementable::provideSupplement):
      (WebCore::Supplementable::removeSupplement):
      (WebCore::Supplementable::requireSupplement):
      (Supplementable):
      (WebCore::Supplementable::Supplementable):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140580 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      570be473
    • jschuh@chromium.org's avatar
      [CHROMIUM] Suppress c4267 build warnings in WTF for Win64 targets · 28afa0f0
      jschuh@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107721
      
      Reviewed by Abhishek Arya.
      
      * WTF.gyp/WTF.gyp:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140579 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      28afa0f0
    • pilgrim@chromium.org's avatar
      [Chromium] Fix some #includes to headers in chromium/public/platform/ that... · 7d52710f
      pilgrim@chromium.org authored
      [Chromium] Fix some #includes to headers in chromium/public/platform/ that have been moved to the new Platform directory
      https://bugs.webkit.org/show_bug.cgi?id=107722
      
      Reviewed by Darin Fisher.
      
      These headers just redirect to the new versions, so call the new
      versions directly.
      
      * src/linux/WebFontInfo.cpp:
      * src/mac/WebSubstringUtil.mm:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140578 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7d52710f
    • oliver@apple.com's avatar
      Add RELEASE_ASSERT (and a few friends) · 1682c0f6
      oliver@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=107725
      
      Reviewed by Filip Pizlo.
      
      Adding a few macros to make release mode assertions easier and more readable.
      Also makes same assertions more useful in debug builds.
      
      * wtf/Assertions.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140577 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1682c0f6
    • tony@chromium.org's avatar
      Incorrect scrollable height during simplified layout · c65268d5
      tony@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107193
      
      Reviewed by David Hyatt.
      
      Source/WebCore:
      
      When computing overflow we need the height of the block before
      it was clamped (i.e., before updateLogicalHeight() has been called).
      
      During simplified layout, we don't have this information and we were
      using the clamped height by mistake. To fix this, we now store the
      pre-clamped height on RenderOverflow so we can properly compute
      overflow.
      
      Test: fast/overflow/height-during-simplified-layout.html
      
      * rendering/RenderBlock.cpp:
      (WebCore::RenderBlock::computeOverflow): Save the height if we have overflow.
      (WebCore::RenderBlock::simplifiedLayout): If we have overflow, use the height that we saved
      in computeOverflow.
      * rendering/RenderOverflow.h:
      (WebCore::RenderOverflow::layoutClientAfterEdge):
      (WebCore::RenderOverflow::setLayoutClientAfterEdge):
      (RenderOverflow): Add a member variable to save the height.
      
      LayoutTests:
      
      * fast/overflow/height-during-simplified-layout-expected.txt: Added.
      * fast/overflow/height-during-simplified-layout.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140576 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c65268d5
    • tsepez@chromium.org's avatar
      [chromium] harden ScriptWrappable::m_wrapper against tampering · f1550185
      tsepez@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107318
      
      Reviewed by Adam Barth.
      
      Patch is correct if existing tests past without crashing.
      
      * bindings/v8/ScriptWrappable.h:
      (WebCore::ScriptWrappable::ScriptWrappable):
      (WebCore::ScriptWrappable::wrapper):
      (WebCore::ScriptWrappable::setWrapper):
      (WebCore::ScriptWrappable::clearWrapper):
      (WebCore::ScriptWrappable::disposeWrapper):
      (WebCore::ScriptWrappable::reportMemoryUsage):
      (ScriptWrappable):
      (WebCore::ScriptWrappable::maskOrUnmaskPointer):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140575 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f1550185
    • roger_fong@apple.com's avatar
      WebCore property sheets, modified build scripts, and project files for compiling in VS2010. · 2d59f5ef
      roger_fong@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=106988
      
      Reviewed by Brent Fulgham.
      
      * WebCore.vcproj/WebCore.sln:
      * WebCore.vcxproj: Added.
      * WebCore.vcxproj/MigrateScripts: Added.
      * WebCore.vcxproj/QTMovieWin: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWin.vcxproj: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWin.vcxproj.filters: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWin.vcxproj.user: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWinCommon.props: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWinDebug.props: Added.
      * WebCore.vcxproj/QTMovieWin/QTMovieWinRelease.props: Added.
      * WebCore.vcxproj/WebCore.vcxproj: Added.
      * WebCore.vcxproj/WebCore.vcxproj.filters: Added.
      * WebCore.vcxproj/WebCore.vcxproj.user: Added.
      * WebCore.vcxproj/WebCoreCFNetwork.props: Added.
      * WebCore.vcxproj/WebCoreCG.props: Added.
      * WebCore.vcxproj/WebCoreCommon.props: Added.
      * WebCore.vcxproj/WebCoreDebug.props: Added.
      * WebCore.vcxproj/WebCoreGenerated.make: Added.
      * WebCore.vcxproj/WebCoreGenerated.vcxproj: Added.
      * WebCore.vcxproj/WebCoreGenerated.vcxproj.filters: Added.
      * WebCore.vcxproj/WebCoreGenerated.vcxproj.user: Added.
      * WebCore.vcxproj/WebCoreGeneratedCommon.props: Added.
      * WebCore.vcxproj/WebCoreGeneratedDebug.props: Added.
      * WebCore.vcxproj/WebCoreGeneratedRelease.props: Added.
      * WebCore.vcxproj/WebCoreMediaQT.props: Added.
      * WebCore.vcxproj/WebCorePostBuild.cmd: Added.
      * WebCore.vcxproj/WebCorePreBuild.cmd: Added.
      * WebCore.vcxproj/WebCorePreLink.cmd: Added.
      * WebCore.vcxproj/WebCorePthreads.props: Added.
      * WebCore.vcxproj/WebCoreQuartzCore.props: Added.
      * WebCore.vcxproj/WebCoreRelease.props: Added.
      * WebCore.vcxproj/build-generated-files.sh: Added.
      * WebCore.vcxproj/copyForwardingHeaders.cmd: Added.
      * WebCore.vcxproj/copyWebCoreResourceFiles.cmd: Added.
      * WebCore.vcxproj/migrate-scripts.sh: Added.
      * WebCore.vcxproj/xcopy.excludes: Added.
      * WebCorePrefix.h:
      * config.h:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140574 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2d59f5ef
    • eric@webkit.org's avatar
      HTMLCompactToken needs to include the forceQuirks bool from HTMLToken · ede568f2
      eric@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107713
      
      Reviewed by Tony Gentilcore.
      
      Before we were only getting "quirks mode" when the system/public identifiers
      were known quirks identifiers.  Now we'll correctly get quirks mode for any
      parse error during DOCTYPE parsing.
      This passes a bunch more tests.
      
      * html/parser/CompactHTMLToken.cpp:
      (WebCore::CompactHTMLToken::CompactHTMLToken):
      * html/parser/CompactHTMLToken.h:
      (WebCore::CompactHTMLToken::doctypeForcesQuirks):
      (CompactHTMLToken):
      * html/parser/HTMLToken.h:
      (WebCore::AtomicHTMLToken::AtomicHTMLToken):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140573 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ede568f2
    • beidson@apple.com's avatar
      Recursion handling cancelled authentication challenges in NetworkProcess · 44925771
      beidson@apple.com authored
      <rdar://problem/13024541> and https://bugs.webkit.org/show_bug.cgi?id=107702
      
      Reviewed by Alexey Proskuryakov.
      
      Source/WebCore:
      
      * WebCore.exp.in:
      
      Source/WebKit2:
      
      This turned in to both a bug fix with authentication and a minimal refactoring of NetworkResourceLoader.
      
      - Rename ::stop to ::resourceHandleStopped
      - Move all cleanup code to ::resourceHandleStopped
      - Schedule a resourceHandleStopped call when an authentication cancellation occurs
      - Tell the WebResourceLoader to cancel when an authentication cancellation occurs
      
      * NetworkProcess/NetworkResourceLoader.cpp:
      (WebKit::NetworkResourceLoader::~NetworkResourceLoader):
      (WebKit::NetworkResourceLoader::start):
      (WebKit::NetworkResourceLoader::performStops):
      (WebKit::NetworkResourceLoader::resourceHandleStopped):
      (WebKit::NetworkResourceLoader::didFail):
      (WebKit::NetworkResourceLoader::receivedAuthenticationCancellation):
      * NetworkProcess/NetworkResourceLoader.h:
      
      * WebProcess/Network/WebResourceLoader.cpp:
      (WebKit::WebResourceLoader::cancelResourceLoader):
      * WebProcess/Network/WebResourceLoader.h:
      * WebProcess/Network/WebResourceLoader.messages.in:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140572 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      44925771
    • bdakin@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=107628 · c975f556
      bdakin@apple.com authored
      Sometimes scroll position is jerky during rubber-band, affects nytimes.com
      -and corresponding-
      <rdar://problem/12679549>
      
      Reviewed by Simon Fraser.
      
      The basic problem here is that isRubberBandInProgress() was only implemented for 
      main thread scrolling. So when we were actually scrolling on the scrolling thread, 
      that function would always return false regardless.
      
      New ScrollableArea virtual function isRubberBandInProgress() will allow us to ask 
      the ScrollingCoordinator when the scrolling thread is scrolling, or the 
      ScrollAnimator otherwise.
      * page/FrameView.cpp:
      (WebCore::FrameView::isRubberBandInProgress):
      * page/FrameView.h:
      (FrameView):
      * platform/ScrollableArea.h:
      (WebCore::ScrollableArea::isRubberBandInProgress):
      
      New ScrollingCoordinator function isRubberBandInProgress() always returns false 
      for non-Mac ports, and is overridden in ScrollingCoordinatorMac to consult the 
      ScrollingTree.
      * page/scrolling/ScrollingCoordinator.h:
      (WebCore::ScrollingCoordinator::isRubberBandInProgress):
      * page/scrolling/mac/ScrollingCoordinatorMac.h:
      (ScrollingCoordinatorMac):
      * page/scrolling/mac/ScrollingCoordinatorMac.mm:
      (WebCore::ScrollingCoordinatorMac::isRubberBandInProgress):
      
      New variable m_mainFrameIsRubberBanding keeps track of whether there is currently 
      a rubber-band happening on the scrolling thread.
      * page/scrolling/ScrollingTree.cpp:
      (WebCore::ScrollingTree::ScrollingTree):
      (WebCore::ScrollingTree::isRubberBandInProgress):
      (WebCore::ScrollingTree::setMainFrameIsRubberBanding):
      * page/scrolling/ScrollingTree.h:
      (ScrollingTree):
      (WebCore::ScrollingTree::rootNode):
      
      Call setMainFrameIsRubberBanding() whenever the stretchAmount is calculated and 
      whenever we stop the rubber-band timer.
      * page/scrolling/mac/ScrollingTreeScrollingNodeMac.mm:
      (WebCore::ScrollingTreeScrollingNodeMac::stretchAmount):
      (WebCore::ScrollingTreeScrollingNodeMac::stopSnapRubberbandTimer):
      
      Consult FrameView for isRubberBandInProgress().
      * platform/ScrollView.cpp:
      (WebCore::ScrollView::updateScrollbars):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140571 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c975f556
    • robert@webkit.org's avatar
      Abspos Inline block not positioned correctly in text-aligned container · 83f864f8
      robert@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=105695
      
      Reviewed by Ojan Vafai.
      
      Source/WebCore:
      
      Inline positioned elements should still obey the text-alignment of their container
      even when we don't do a line layout.
      
      Test: fast/text/container-align-with-inlines.html
      
      * rendering/RenderBlock.cpp:
      (WebCore::RenderBlock::adjustPositionedBlock):
      (WebCore::RenderBlock::updateStaticInlinePositionForChild):
      (WebCore):
      * rendering/RenderBlock.h:
      (RenderBlock):
      * rendering/RenderBlockLineLayout.cpp:
      (WebCore::setStaticPositions):
      
      LayoutTests:
      
      * fast/text/container-align-with-inlines-expected.txt: Added.
      * fast/text/container-align-with-inlines.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140570 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      83f864f8
    • eric@webkit.org's avatar
      Remove DocType support from MarkupTokenBase now that NEW_XML is gone · 3dba811f
      eric@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107709
      
      Reviewed by Adam Barth.
      
      Just moving code out of MarkupTokenBase and into HTMLToken.
      The test for this change is if it compiles. :)  Which it does.
      
      * html/parser/HTMLToken.h:
      (DoctypeData):
      (WebCore::DoctypeData::DoctypeData):
      (WebCore):
      (WebCore::HTMLToken::appendToName):
      (WebCore::HTMLToken::name):
      (HTMLToken):
      (WebCore::HTMLToken::beginDOCTYPE):
      (WebCore::HTMLToken::publicIdentifier):
      (WebCore::HTMLToken::systemIdentifier):
      (WebCore::HTMLToken::setPublicIdentifierToEmptyString):
      (WebCore::HTMLToken::setSystemIdentifierToEmptyString):
      (WebCore::HTMLToken::appendToPublicIdentifier):
      (WebCore::HTMLToken::appendToSystemIdentifier):
      (WebCore::HTMLToken::releaseDoctypeData):
      (WebCore::AtomicHTMLToken::AtomicHTMLToken):
      (AtomicHTMLToken):
      * html/parser/HTMLTokenTypes.h:
      * xml/parser/MarkupTokenBase.h:
      (WebCore):
      (MarkupTokenBase):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140569 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      3dba811f
    • roger_fong@apple.com's avatar
      Add the VS2010 WebKit solution folder and solution file. · b637c8b7
      roger_fong@apple.com authored
      The solution of course will not load until the dependent projects are all completed.
      https://bugs.webkit.org/show_bug.cgi?id=106949.
      
      Reviewed by Brent Fulgham.
      
      * WebKit.vcxproj: Added.
      * WebKit.vcxproj/WebKit.sln: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140568 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b637c8b7
    • roger_fong@apple.com's avatar
      Add WebKitLibraries property sheets for compiling WebKit in VS2010. · 5e0bbf62
      roger_fong@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=106949
      
      Reviewed by Brent Fulgham.
      
      * win/tools/vsprops/FeatureDefines.props: Added.
      * win/tools/vsprops/common.props: Added.
      * win/tools/vsprops/debug.props: Added.
      * win/tools/vsprops/release.props: Added.
      * win/tools/vsprops/releaseproduction.props: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140567 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      5e0bbf62
    • pilgrim@chromium.org's avatar
      [Chromium] Fix some #includes for WebSerializedScriptValue · 8072fb67
      pilgrim@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107697
      
      Reviewed by Adam Barth.
      
      WebSerializedScriptValue has been moved to chromium/public/.
      
      * public/WebDOMCustomEvent.h:
      * public/WebDOMMessageEvent.h:
      * public/WebIDBCursor.h:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140566 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8072fb67
    • pilgrim@chromium.org's avatar
      [Chromium] Update some #includes in DumpRenderTree for headers in the new Platform directory · 213c0a7d
      pilgrim@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107145
      
      Reviewed by Adam Barth.
      
      Part of a larger refactoring series; see tracking bug 82948.
      
      * DumpRenderTree/chromium/DRTDevToolsAgent.cpp:
      * DumpRenderTree/chromium/DRTDevToolsAgent.h:
      * DumpRenderTree/chromium/DRTDevToolsClient.h:
      * DumpRenderTree/chromium/DRTTestRunner.cpp:
      * DumpRenderTree/chromium/DRTTestRunner.h:
      * DumpRenderTree/chromium/MockWebPrerenderingSupport.h:
      * DumpRenderTree/chromium/MockWebSpeechInputController.cpp:
      * DumpRenderTree/chromium/MockWebSpeechInputController.h:
      * DumpRenderTree/chromium/NotificationPresenter.cpp:
      * DumpRenderTree/chromium/TestNavigationController.h:
      * DumpRenderTree/chromium/TestRunner/src/AccessibilityControllerChromium.cpp:
      * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
      * DumpRenderTree/chromium/TestRunner/src/CppBoundClass.cpp:
      * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
      (WebTestRunner):
      * DumpRenderTree/chromium/TestRunner/src/EventSender.h:
      * DumpRenderTree/chromium/TestRunner/src/GamepadController.h:
      * DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
      * DumpRenderTree/chromium/TestRunner/src/TestInterfaces.cpp:
      * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
      * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
      * DumpRenderTree/chromium/TestRunner/src/TextInputController.cpp:
      * DumpRenderTree/chromium/TestRunner/src/WebPermissions.cpp:
      * DumpRenderTree/chromium/TestRunner/src/WebTestPlugin.cpp:
      * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
      * DumpRenderTree/chromium/TestShell.cpp:
      * DumpRenderTree/chromium/WebThemeEngineDRTMac.h:
      * DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
      * DumpRenderTree/chromium/WebThemeEngineDRTWin.cpp:
      * DumpRenderTree/chromium/WebThemeEngineDRTWin.h:
      * DumpRenderTree/chromium/WebViewHost.cpp:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140565 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      213c0a7d
    • nghanavatian@rim.com's avatar
      [BlackBerry] Take a better guess as to what text the user wants learned · 55af1a7c
      nghanavatian@rim.com authored
      https://bugs.webkit.org/show_bug.cgi?id=107706
      
      Reviewed by Rob Buis.
      
      If we still have a valid caret position, use it to give priority to a region
      of text.  We now take the text before the caret, instead of the text at the end
      of the field.  If the caret is not present, we start from the beginning of the field.
      This helps learning in email where the region of interest is generally at the start,
      not end of the field.
      
      Internally reviewed by Mike Fenton
      
      * WebKitSupport/InputHandler.cpp:
      (BlackBerry::WebKit::InputHandler::learnText):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140564 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      55af1a7c
    • zandobersek@gmail.com's avatar
      Unreviewed. · 9dbe1cc8
      zandobersek@gmail.com authored
      Speculative build fix for the GTK port after 140539.
      
      * Source/autotools/symbols.filter:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140563 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9dbe1cc8
    • nghanavatian@rim.com's avatar
      [BlackBerry] Optimize spellchecking by coalescing messages · 31cc0900
      nghanavatian@rim.com authored
      https://bugs.webkit.org/show_bug.cgi?id=107707
      
      Reviewed by Rob Buis.
      
      PR233604
      Instead of taking chunks one line at a time, coalesce them together
      to fire off messages as close to our character limit as possible.
      This should dramatically reduce the total number of messages in email
      giving us a little performance bump.
      
      Internally reviewed by Mike Fenton and Gen Mak.
      
      * WebKitSupport/DOMSupport.cpp:
      (BlackBerry::WebKit::DOMSupport::trimWhitespaceFromRange):
      (DOMSupport):
      * WebKitSupport/DOMSupport.h:
      * WebKitSupport/InputHandler.cpp:
      * WebKitSupport/InputHandler.h:
      (InputHandler):
      * WebKitSupport/SpellingHandler.cpp:
      (BlackBerry::WebKit::SpellingHandler::spellCheckTextBlock):
      (BlackBerry::WebKit::SpellingHandler::createSpellCheckRequest):
      (BlackBerry::WebKit::SpellingHandler::parseBlockForSpellChecking):
      (BlackBerry::WebKit::SpellingHandler::getRangeForSpellCheckWithFineGranularity):
      * WebKitSupport/SpellingHandler.h:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140562 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      31cc0900
    • jochen@chromium.org's avatar
      [chromium] Use after free in plugins/geturlnotify-during-document-teardown.html · 481a49d6
      jochen@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107556
      
      Reviewed by Tony Chang.
      
      WebViewHost initiates a navigation to about:blank in its destructor.
      However, since WebTestProxy inherits from WebViewHost, at this point
      the WebViewClient and WebFrameClient interfaces are already partially
      destructed resulting in the use after free.
      
      This does not affect the chromium implementation since it doesn't
      invoke WebKit API methods in its destructor.
      
      * DumpRenderTree/chromium/TestShell.cpp:
      (TestShell::~TestShell):
      (TestShell::closeWindow):
      * DumpRenderTree/chromium/WebViewHost.cpp:
      (WebViewHost::WebViewHost):
      (WebViewHost::~WebViewHost):
      (WebViewHost::shutdown):
      * DumpRenderTree/chromium/WebViewHost.h:
      (WebViewHost):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140561 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      481a49d6
    • alexis@webkit.org's avatar
      transition-property accepts incorrect "all, none" as value · c2303abf
      alexis@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=105428
      
      Reviewed by Dean Jackson.
      
      Source/WebCore:
      
      http://www.w3.org/TR/css3-transitions/#transition-property-property
      disallows any value like none, all or all, none as it doesn't make
      sense. This patch fixes the problem by rejecting the value if set by
      the user.
      
      Test: transitions/transitions-parsing.html
      
      * css/CSSParser.cpp:
      (WebCore::CSSParser::parseAnimationProperty):
      * css/CSSParser.h:
      
      LayoutTests:
      
      Extend the existing tests to cover the bug.
      
      * transitions/transitions-parsing-expected.txt:
      * transitions/transitions-parsing.html:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140560 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c2303abf
    • eric@webkit.org's avatar
      Clarify some usage of shouldUseThreading vs m_haveBackgroundParser, fixing about:blank parsing · 209109c8
      eric@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107664
      
      Reviewed by Adam Barth.
      
      The HTMLDocumentParser can be "supposed" to use a background parser
      but not end up ever doing so for blank documents.  It's important
      that we spin the forground parser at least once, so we end up
      actually creating the about:blank document.
      
      shouldUseThreading() means that threading is enabled and we should use it if we can.
      m_haveBackgroundParser means that we actually are already using threading.
      When we add full document.write support this will get a bit trickier still as
      we'll need to be using both the foreground and background parsers.
      
      This fixes about 20 tests in fast/frames and many others outside of fast/frames. :)
      
      * html/parser/HTMLDocumentParser.cpp:
      (WebCore::HTMLDocumentParser::prepareToStopParsing):
      (WebCore::HTMLDocumentParser::resumeParsingAfterYield):
      (WebCore::HTMLDocumentParser::finish):
      (WebCore::HTMLDocumentParser::lineNumber):
      (WebCore::HTMLDocumentParser::textPosition):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140559 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      209109c8
    • roger_fong@apple.com's avatar
      [Win] Remove dependence on Microsoft Embedded OpenType Font Engine... · 70f4208c
      roger_fong@apple.com authored
      [Win] Remove dependence on Microsoft Embedded OpenType Font Engine (T2EMBED.DLL)  from FontCustomPlatformData.cpp.
      https://bugs.webkit.org/show_bug.cgi?id=107153
      
      Reviewed by Dan Bernstein.
      
      * platform/graphics/win/FontCustomPlatformData.cpp:
      (WebCore):
      (WebCore::FontCustomPlatformData::~FontCustomPlatformData):
      (WebCore::FontCustomPlatformData::fontPlatformData):
      (WebCore::createFontCustomPlatformData):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140558 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      70f4208c
    • commit-queue@webkit.org's avatar
      [GTK] Avoid reset title for navigation within the page · f54743a8
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=106908
      
      Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-01-23
      Reviewed by Martin Robinson.
      
      * WebCoreSupport/FrameLoaderClientGtk.cpp:
      (WebKit::FrameLoaderClient::dispatchDidNavigateWithinPage): Call
      dispatchDidCommitLoad with true as param.
      (WebKit::FrameLoaderClient::dispatchDidCommitLoad): The method has been
      overloaded. The default implementation (without params) simply calls
      dispatchDidCommitLoad with false. The new private method with
      isNavigatingWithinPage as param will avoid to reset the title for
      navigation within the page.
      * WebCoreSupport/FrameLoaderClientGtk.h:
      (FrameLoaderClient): Add new private method dispatchDidCommitLoad with a
      boolean parameter to know if it is navigating withing the same page or
      not.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140557 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f54743a8
    • commit-queue@webkit.org's avatar
      Web Inspector: speedup highlight regex API in DefaultTextEditor · 4320567c
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107238
      
      Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-23
      Reviewed by Pavel Feldman.
      
      Source/WebCore:
      
      Move overlay highlight measurement (highlight regex API) from DefaultTextEditor._paintLine
      method to the DefaultTextEditor._paintLines method which allows to relayout dom only
      once. This is a significant improvement to the current state of the
      art which does relayout on each regex occurence.
      In addition, use "left" css attribute instead of "margin-left": this
      way it will be possible to avoid an unnecessary relayouting during
      appending overlay highlight.
      
      No new tests: no change in behaviour.
      
      * inspector/front-end/DefaultTextEditor.js:
      (WebInspector.TextEditorMainPanel.prototype.markAndRevealRange):
      (WebInspector.TextEditorMainPanel.prototype._paintLines):
      (WebInspector.TextEditorMainPanel.prototype._measureRegexHighlight):
      (WebInspector.TextEditorMainPanel.prototype._measureSpans):
      (WebInspector.TextEditorMainPanel.prototype._appendOverlayHighlight):
      (WebInspector.TextEditorMainPanel.prototype._paintLine):
      (WebInspector.TextEditorMainPanel.ElementMetrics):
      (WebInspector.TextEditorMainPanel.LineOverlayHighlight):
      (WebInspector.TextEditorMainChunk.prototype.expand):
      
      LayoutTests:
      
      Corrected layout test expectations to correspond to refactoring.
      
      * inspector/editor/text-editor-highlight-regexp-expected.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140556 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      4320567c
    • ojan@chromium.org's avatar
      Assert that computePreferredLogicalWidths never calls setNeedsLayout · 29795f57
      ojan@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107613
      
      Reviewed by Eric Seidel.
      
      computePreferredLogicalWidths should only set m_minPreferredLogicalWidth
      and m_maxPreferredLogicalWidth. It shouldn't have other side-effects.
      The mathml bits can be removed once https://bugs.webkit.org/show_bug.cgi?id=107353
      is resolved.
      
      * rendering/RenderBox.cpp:
      (WebCore::RenderBox::minPreferredLogicalWidth):
      * rendering/mathml/RenderMathMLOperator.cpp:
      (WebCore::RenderMathMLOperator::computePreferredLogicalWidths):
      * rendering/mathml/RenderMathMLRoot.cpp:
      (WebCore::RenderMathMLRoot::computePreferredLogicalWidths):
      * rendering/mathml/RenderMathMLRow.cpp:
      (WebCore::RenderMathMLRow::computePreferredLogicalWidths):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140554 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      29795f57
    • pfeldman@chromium.org's avatar
      Follow up to 140539, adding exports for Win build. · 7b8ba498
      pfeldman@chromium.org authored
      Not reviewed.
      
      * WebKit.vcproj/WebKitExports.def.in:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140553 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7b8ba498
    • commit-queue@webkit.org's avatar
      [GTK][WK2] Unflag tests related to layoutTestController.setViewModeMediaFeature · a6aa7bb4
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107660
      
      Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-01-23
      Reviewed by Philippe Normand.
      
      * platform/gtk-wk2/TestExpectations: Remove tests related to
      layoutTestController.setViewModeMediaFeature as they are passing now.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140550 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a6aa7bb4
    • commit-queue@webkit.org's avatar
      REGRESSION(r140518): Broke Chromium Win build (gyp file not updated) (Requested by arv on #webkit). · 9a73d970
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=107691
      
      Unreviewed build fix. Removed TextureMapperShaderManager.* from WebCore.gypi.
      
      Patch by Noam Rosenthal <noam@webkit.org> on 2013-01-23
      
      * WebCore.gypi:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@140547 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9a73d970